fig3

Figure 3. Development of the m6A-based gene signature. (A) Univariable Cox regression analyses measuring the predictive value of m6A methylation regulators for overall survival of ccRCC patients. (B) Tuning parameter (λ) selection in the LASSO model used 10-fold cross-validation via minimum criteria. Partial likelihood deviances from the LASSO regression cross-validation procedure were drawn as a function of log(λ). The numbers along the upper x-axis represent the average number of predictors. The red dots show the average deviance values for each model with a given λ, and the vertical bars through the red dots indicate the upper and lower values of the deviance. The dotted vertical lines are plotted at the optimal values where the model provides its best fit to the data. The optimal λ value of 0.038 with log (λ) = -3.279 was identified. (C) LASSO coefficient profiles of the 23 m6A methylation regulators. The dotted vertical line is plotted at the value determined by using 10-fold cross-validation in (B), where optimal λ resulted in 5 nonzero coefficients. (D) Histogram showing the coefficients of the selected m6A regulators in the gene signature. ccRCC: Clear cell renal cell carcinoma; LASSO: least absolute shrinkage and selection operator.